This revised application proposes to focus on the role of adenosine Al receptors (A1AR) in the preconditioning phenomenon. Five min of transient coronary occlusion (preconditioning) renders the heart very resistant to infarction from a subsequent ischemic event. Since the last submission, we have found that the protection afforded by a preconditioning period of ischemia can be blocked by adenosine receptor antagonists suggesting that the build up of adenosine mediates the protection. Further support for that theory was provided by the observation that a transient infusion of intracoronary adenosine or an A1AR selective agonist, PIA, could be substituted for the ischemia and confer the same protection. The PIA result indicates that A1AR mediate the response. PIA, but not adenosine, was also found to be A1AR selective enough that it could confer protection even when infused IV. Studies are proposed to further test this theory. We will test whether PIA can induce a sustained reduction of infarct size in the rabbit model. Adenosine and PIA will also be tested in a canine model of -stunning and a rat model of reperfusion arrhythmias to see if A1ARinduced protection extends to these other forms of cardiac injury. We will test whether the heart can be maintained in a preconditioned state pharmacologically over extended periods of time. We will test whether augmented glycolytic capacity could be mechanism for A1AR induced protection. Although PIA is a powerful protectant, it is also mutagenic and unsuitable for human use. We will, therefore, test a panel of A1AR selective agents which should not be mutagenic in the infarct size model. Finally we will test whether Gi proteins are involved in this response. The overall aim is to identify a compound which would be capable of maintaining a patient's heart in a preconditioned state indefinitely.